linux-kernel - Re: WARN_ON_ONCE(!new_owner) within wake_futex

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <alpine.DEB.2.21.1901291012330.1669@nanos.tec.linutronix.de>
Date:   Tue, 29 Jan 2019 10:45:44 +0100 (CET)
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Heiko Carstens <heiko.carstens@...ibm.com>
cc:     Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...nel.org>,
        Martin Schwidefsky <schwidefsky@...ibm.com>,
        LKML <linux-kernel@...r.kernel.org>, linux-s390@...r.kernel.org,
        Stefan Liebler <stli@...ux.ibm.com>,
        Sebastian Sewior <bigeasy@...utronix.de>
Subject: Re: WARN_ON_ONCE(!new_owner) within wake_futex_pi() triggered

On Tue, 29 Jan 2019, Heiko Carstens wrote:
> On Mon, Jan 28, 2019 at 04:53:19PM +0100, Thomas Gleixner wrote:
> > Patch below cures that.
> 
> With your patch the kernel warning doesn't occur anymore. So if this
> is supposed to be the fix feel free to add:

Yes, it's supposed to be the fix.
> 
> However now I see every now and then the following failure from the
> same test case:
> 
> tst-robustpi8: ../nptl/pthread_mutex_lock.c:425: __pthread_mutex_lock_full: Assertion `INTERNAL_SYSCALL_ERRNO (e, __err) != ESRCH || !robust' failed.
> 
> 		/* ESRCH can happen only for non-robust PI mutexes where
> 		   the owner of the lock died.	*/
> 		assert (INTERNAL_SYSCALL_ERRNO (e, __err) != ESRCH || !robust);
> 
> I just verified that this happened also without your patch, I just
> didn't see it since I started my tests with panic_on_warn=1 and the
> warning triggered always earlier.
> So, this seems to be something different.

Moo. I ran the test loop all night (simply because I forgot to stop it) and
of course this does not trigger here. Could you try to gather a bit more
information with lightweight tracing?

Something like the below should give us at least a clue. Stop the tst loop
when the assert triggers and then extract the trace.

Thanks,

	tglx

8<--------------

#!/bin/sh

# Substitute with your ld comm string if it starts differently 
C=ld-linux

echo 'comm ~ "$C*"' >/sys/kernel/debug/tracing/events/syscalls/sys_enter_futex/filter
echo 'comm ~ "$C*"' >/sys/kernel/debug/tracing/events/syscalls/sys_exit_futex/filter
echo 'comm ~ "$C*"' >/sys/kernel/debug/tracing/events/sched/sched_process_exit/filter
echo 'prev_comm ~ "$C*" || next_comm ~ "$C*"' >/sys/kernel/debug/tracing/events/sched/sched_switch/filter

echo 1 > /sys/kernel/debug/tracing/events/syscalls/sys_enter_futex/enable
echo 1 > /sys/kernel/debug/tracing/events/syscalls/sys_exit_futex/enable
echo 1 > /sys/kernel/debug/tracing/events/sched/sched_process_exit/enable
echo 1 > /sys/kernel/debug/tracing/events/sched/sched_switch/enable